We assessed the extent to which parents of children with autism spectrum disorder report that they are engaged in shared decision making. We measured the association between shared decision making and (a) satisfaction with care, (b) perceived guidance regarding controversial issues in autism spectrum disorder, and (c) perceived assistance navigating the multitude of treatment options. Surveys assessing primary medical care and decision-making processes were developed on the basis of the U.S. Department of Health and Human Service's Consumer Assessment of Healthcare Providers and Systems survey. In May 2009, after pilot testing, we sent surveys to 203 parents of children from ages 3 to 18 with International Classification of Diseases–9 and parent-confirmed autism spectrum disorder diagnoses. The response rate was 64%. Controlling for key demographic variables, parents of children with autism spectrum disorder reporting higher levels of shared decision making reported significantly greater satisfaction with the overall quality of their child's health care (p ≤ .0001). Parents reporting higher levels of shared decision making were also significantly more likely to report receiving guidance on the many treatment options (p = .0002) and controversial issues related to autism spectrum disorder (p = .0322). In this study, shared decision making was associated with higher parent satisfaction and improved guidance regarding treatments and controversial issues within primary care for children with autism spectrum disorder.
Primary medical care for children with autism spectrum disorder (ASD) is complex because children with ASD present with a broad range of symptoms, treatments for ASD often lack supportive evidence, and parent–provider relationships have involved a history of controversy. Such challenges may be one of many factors contributing to the fact that when providing care for children with ASD, primary care physicians have reported feeling less competent and perceived that parents are less likely to trust them than when providing care for children with other neurodevelopmental and chronic conditions (Golnik, Ireland, & Borowsky, 2009). These physician challenges are mirrored by caregivers (henceforth referred to as parents) of children with ASD, who rate physicians lower on their ability to answer questions about the child's condition and to understand the condition's impact on the family than do parents of children with other intellectual or physical disabilities (Liptak et al., 2006). Furthermore, parents of children with ASD have reported that their children are less likely to receive primary care consistent with the American Academy of Pediatrics, Ad Hoc Task Force on Definition of the Medical Home's (1992) medical home guidelines, such as family-centered care (Brachlow, Ness, McPheeters, & Gurney, 2007; Kogan et al., 2008; Strickland et al., 2004), and more likely to experience unmet health care needs (Kogan et al., 2008) than children with other types of special health care needs. Considering the numerous autism treatment options and the history of controversy regarding medical issues, we propose that shared decision making (SDM) may facilitate positive primary care relationships and improved medical care for children with ASD.
The degree and nature of presenting symptoms for an ASD diagnosis (qualitative impairments in social communication and restricted, repetitive behaviors, interests, and activities) are highly variable (American Psychiatric Association, 1994). Each unique child with ASD uses a vast and individualized array of changing and evolving therapies and treatments (e.g., intensive behavioral intervention, speech therapy, occupational therapy, educational supports; Myers & Johnson 2007). Additional treatments are often used to address diverse, commonly co-occurring issues (e.g., irritability, inattention, gastrointestinal issues, seizures, insomnia; Buie et al., 2010; Myers & Johnson, 2007).
Currently no single, standard medical or biological treatment exists for the core features of ASD (Lord et al., 2005). In addition to the wide array of treatments, many treatments lack conclusive evidence. Although forms of intensive behavioral intervention are clearly associated with some improvements in language, IQ, academic functioning, and social and adaptive behavior and reductions in repetitive and disruptive behaviors, psychosocial and educational interventions have varying efficacy depending on intervention and child factors (Lord & Bishop, 2010; National Autism Center, 2009). Furthermore, more than half of families of children with ASD use complementary alternative medicine treatments (Hanson, Kalish, Merhar, Ittenbach, & Pinto-Martin, 2007; Harrington, Rosen, Gamecho, & Patrick, 2006), which are supported by wide-ranging degrees of evidence and have potential side effects. Although etiological theories such as immune system (Ashwood, Wills, & Van de Water, 2006) or gastrointestinal dysregulation (Buie et al., 2010) are actively being researched, their clinical application in the general primary care setting is frequently unclear.
Historically, clinicians and parents have had some disagreement regarding the etiology and treatment of autism (Silverman & Brosco, 2007). Tensions arose when Kanner in the 1940s and Bettelheim in the 1960s incorrectly attributed ASD to parental emotional coldness (Olley & Gutentag, 1999) or, more recently, when some parents withheld vaccinations out of concern that they were associated with ASD (Freed, Clark, Butchart, Singer, & Davis, 2010). Parents of children with ASD often obtain ASD information from the media and other parents (Rhoades, Scarpa, & Salley, 2007). Such advice may differ from recommendations based on medical evidence.
SDM may be particularly well suited to primary care of children with ASD. Rather than providing a treatment plan based on the condition, a physician using SDM engages in a dialogue, inquiring about the family's preferences, values, and quality of life, as well as the child's symptoms. The physician offers information regarding the potential outcomes, risks, and benefits of a treatment option. Then the physician and family jointly participate in developing and agree on a treatment plan for the specific health issues of that unique child (Charles & Whelan, 1997). SDM is an alternative to both the paternalistic model, in which “doctor knows best” and the patient merely gives consent, and the informed decision-making model, in which the patient alone makes the decision after being given all the technical information (Charles & Whelan, 1997). SDM is most appropriate with decisions of low certainty (Whitney 2003), when no one clear best treatment path is indicated, as is often the case with ASD. SDM is likely to be well suited to families of children with autism, because parents of children with ASD are often very informed and empowered; in the past, they have made substantial contributions to ASD research and political advocacy, predicted paradigmatic shifts in the conceptualization of ASD, and popularized empirically based therapies (Silverman & Brosco, 2007).
Parents of children with ASD have been found to have more stress and adjustment problems than parents of typically developing children and parents of children with other disabilities (Bouma & Schweitzer, 1990; Sanders & Morgan, 1997). Mothers of children with autism have reported less marital satisfaction, less parenting competence, and less family adaptability than mothers of children with Down syndrome or typically developing children (Rodrigue, Morgan, & Geffken, 1990). Challenges specific to parents of children with autism include social isolation because of disruptive child behaviors (e.g., obsessive behaviors, tantrums); difficulties with getting an accurate, timely diagnosis and appropriate treatment; and the lack of a cure for autism (Gray & Holden, 1992). Such studies have underscored the need to increase family support such as parent-to-parent programs, evaluate parental stress levels as part of screening and assessment for ASD interventions, and offer supplementary interventions for parents such as respite care and classes in stress management procedures (Baker-Ericzen, Brookman-Frazee, & Stahmer, 2005). SDM is not merely a means to arrive at the right treatment plan for a child; the process itself has the potential to increase parental satisfaction, empower parents, and aid physicians in identifying parental needs as well.
SDM is a specific tool within the broader conceptual framework of family-centered care, a component of the American Academy of Pediatrics Ad Hoc Task Force on Definition of the Medical Home's (1992) medical home guidelines. Family-centered care involves (a) establishing the family as the focus of services, (b) providing services that strengthen family functioning, and (c) supporting and respecting family decision making (Shelton & Stepanek, 1994). Family-centered practice is at the forefront of contemporary thinking about how early intervention programs, human services, and health care organizations should involve families of children with special health care needs (Dunst, Trivette, & Hamby, 2007), and this general approach has been proven to improve parent empowerment, increase parents' ability to access needed resources, improve parent satisfaction, decrease stress, and heighten emotional well-being (Dempsey & Keen, 2008). However, these studies did not analyze outcomes with regard to specific practices within family-centered care such as SDM.
In other patient populations, engaging patients in decision making has been associated with positive outcomes such as improved satisfaction (Frosch & Kaplan, 1999; Roter & Hall, 1992), health outcomes (Greenfield, Kaplan, Ware, Yano, & Frank, 1988; Kaplan, Greenfield, & Ware, 1989; Stewart 1995), and parents' confidence in managing their child's condition (Clark et al., 1998), along with reduced antibiotic overuse (Merenstein, Diener-West, Krist, Pinnegar, & Cooper, 2005). A study of pediatric patients with attention deficit/hyperactivity disorder and asthma found that children with behavioral impairments or poorer general health were less likely to experience SDM (Fiks, Localio, Alessandrini, Asch, & Guevara, 2010); children with autism have behavioral impairments and may similarly be less likely to experience SDM. Barriers to SDM include lack of time (Gravel, Legare, & Graham, 2006) and potential anxiety that patients may feel when informed of all health care risks (Dawes & Davison 1994). No studies have yet assessed SDM in the medical care of children with ASD.
We undertook this study to determine the extent to which parents of children with ASD report that they are engaged in SDM. We tested the association between SDM and satisfaction with care and perceived guidance regarding controversial ASD issues and navigating the multitude of ASD treatment options. We hypothesized that SDM would be associated with greater satisfaction with health care, perceived guidance regarding controversial ASD issues, and navigation of the multitude of ASD treatment options.
One hundred and thirty of 203 potential respondents completed the survey. Included in the study sample were 128 parents of children ages 0–18 years who received health care in the Fairview Health System's 40 primary care clinics between January 1, 2008, and December 31, 2008, and had an International Classification of Diseases–9 (ICD-9; Center for Disease Control and Prevention, 2009) diagnosis code of ASD (299.00 or 299.01; includes autism, pervasive developmental disorder, and Asperger disorder) in their medical record. This diagnosis was verified by their parents.
Survey Instrument and Administration
We developed an instrument consisting of 13 items regarding demographics and 27 items regarding access, use, and satisfaction with primary medical care and decision-making processes. We conducted pilot testing of the survey with a convenience sample of seven parents of children with ASD to assess clarity and ease of administration. The survey took approximately 10 min to complete and is available on request from Allison Golnik. Surveys were sent to 203 potential participants by U.S. mail in May 2009. Nonrespondents received a second mailing 1–2 months after the first mailing. All participants received and signed an informed consent letter explaining the study's purpose, potential risks and benefits, and contact information. Respondents received a $10 gift card on completion of the survey. The University of Minnesota Institutional Review Board approved the study.
Sociodemographic variables can be found in Table 1. The question that measured functional ability was based on the National Survey of Children with Special Health Care Needs (Child and Adolescent Health Measurement Initiative, 2008). It asked about the frequency and degree to which the child's condition affected his or her ability to do the things done by most other children of the same age, answered on a Likert scale ranging from 1 (no limitations) to 10 (very limited).
Independent Variable: Shared Decision Making
SDM questions were based on the U.S. Department of Health and Human Service's Consumer Assessment of Healthcare Providers and Systems survey (Agency for Healthcare Research and Quality, 2010). The instrument included a Likert scale of agreement ranging from 1 (never) to 7 (always), and parents were asked whether health care staff (a) provided parents with more than one treatment choice, (b) talked with parents about the reasons for and against each choice, (c) gave parents sufficient information about each choice, and (d) asked for parents' participation in health care decision making. Responses to these four questions were summed together to create an overall SDM score for respondents who answered all four questions; a Cronbach's α of more than .9 for this sample's data indicated very strong internal consistency for the overall SDM variable. SDM scores from all respondents were combined to calculate a mean ASD SDM score.
Dependent variables included satisfaction with care and guidance on controversial issues and treatment options. Satisfaction was assessed with the question, “How satisfied are you with the overall quality of health care your child has received?” Likert response categories ranged from 1 (not at all) to 7 (very much so). Guidance regarding controversial issues in ASD and guidance on the many treatment options were assessed with the question, “Rate how well the care team assisted with this issue.” Likert response categories ranged from 1 (not at all) to 5 (very well). Guidance issues were weighted on the basis of level of reported importance; the more important an issue (e.g., nutrition) was rated by the parent on the Likert scale, the more value it was given in the analysis.
Demographic data were assessed for all respondents (Table 1). We used multiple linear regression to measure the association of SDM with parent satisfaction and perceived guidance on controversial issues and treatment options. To control for potential confounding effects, we included in the models the child's age, gender, and functional ability; family income; and insurance type along with respondent type (mother, father) and respondent education. Continuous variables were standardized to a mean of zero and a standard deviation of 1 for the multiple linear regressions. All analyses were conducted using the GENMOD procedure in SAS 9.2 (SAS Institute Inc., Cary, NC).
A total of 130 completed surveys were returned (64%). Two of the 130 surveys were not included because of missing SDM data. Parents reporting higher levels of SDM also reported significantly greater satisfaction with their child's overall quality of health care (p < .0001). Parents reporting greater SDM were also significantly more likely to report receiving more guidance with the many treatment options (p = .0002) and controversial issues related to ASD (p = .0322; Table 2). The R2 values for overall quality of health care, guidance on treatment options, and guidance on controversial issues related to ASD were 0.46, 0.19, and 0.24, respectively. The mean combined SDM score for all respondents was 3.99 (SD = 1.70) on a Likert scale ranging from 1 to 7.
SDM was positively associated with parents of children with ASD feeling satisfied with their child's health care and with perceptions that they had received better guidance on the many treatment options available and controversial issues related to care for children with ASD.
Although other studies have found that SDM is correlated with high patient satisfaction (Frosch & Kaplan, 1999; Merenstein et al., 2005; Roter & Hall, 1992), this study is the first to illustrate the strong positive correlation between SDM and parent satisfaction in primary care of children with ASD. This finding is particularly important considering the history of parental dissatisfaction with ASD medical care (Liptak et al., 2006) and tension between parents of children with ASD and physicians (Freed et al., 2010; Olley & Gutentag, 1999; Silverman & Brosco, 2007). This finding may be significant for health outcomes for children with ASD, because previous studies have found that patient satisfaction is positively correlated with continuity of care (Saultz & Albedaiwi, 2004) and patient compliance (Williams, 1994).
Parents have reported a moderate or a great deal of negative impact on their lives related to difficulty getting treatments (70%), disappointing treatments (46%), or setbacks in the child's treatment progress (70%; Autism Speaks 2010). Furthermore, parents of children with autism often turn to sources other than their primary care physician for treatment information, which can conflict with medical recommendations (Rhoades et al., 2007). Our study highlights the physician's potential to positively assist families who may experience stress related to ASD care or who may turn elsewhere for medical guidance.
The parent–primary care provider relationship can be strained regarding controversial issues, such as when parents choose not to follow recommended immunization schedules, use outside information that conflicts with physician recommendations, or use a complementary alternative medicine modality that carries risks and for which evidence is lacking. Although we did not measure final treatment decisions (e.g., whether families used a potentially risky complementary alternative medicine or did not immunize), our study suggests that through the process of SDM, parents felt that they received better guidance on these potentially contentious issues. SDM could serve as a bridge to facilitate provider–parent partnerships regarding controversial issues frequently encountered in primary care.
Our ability to make causal claims on the basis of these results is limited by the cross-sectional and observational nature of the analysis and the lack of random sampling. Of survey respondents, 66% were college educated. Because patients with higher education are more likely to want to engage in SDM (Strull, Lo, & Charles, 1984), may have more access to ASD resources, and may have better communication skills than parents with less education, the study results could be biased by the respondent population. ASD diagnoses came from medical records confirmed by parental report but were not validated through review of standardized ASD evaluation or direct observation. We were not able to detect whether another factor, such as a family-focused medical home, influenced the relationship between SDM and the outcomes. However, SDM is an inherent part of the family-focused medical home model (American Academy of Pediatrics, 2002), so we hypothesize that any association between the medical home and satisfaction among parents of children with ASD may be partially attributed to SDM.
Future research should evaluate a group randomized to receive SDM compared with a control group of parents with a more passive decision-making role. Observation of physician–family encounters with objective rating of SDM criteria could be used to ensure fidelity. Additional research should test the correlation of SDM in ASD care with continuity of care, family stress directly related to the child's medical care, and compliance with recommendations such as those regarding complementary alternative medicine that carries potential risks and for which evidence is lacking, immunization uptake, and enrollment in evidence-based early intervention services.
This study provides preliminary evidence that SDM is associated with higher parent satisfaction and improved guidance regarding treatments and controversial issues within primary care for children with ASD. SDM, which acknowledges both treatment uncertainty and parental contributions, has the potential to facilitate more consistent, trusted, and effective care for children with ASD. Additional research is needed to determine the potential of purposeful SDM to improve comprehensive, coordinated, and consistent primary care and adherence to treatment recommendations, thus reducing disparities in primary care for children with ASD.
We acknowledge Dana Boergerhoff for her assistance with data collection. Project support was provided by Primary Care Clinic Models of Care Delivery Pilot Projects, Minnesota Department of Human Services, November 1, 2008–October 31, 2010.
Editor-in-Charge: Glenn Fujiura
Allison Golnik (e-mail: Allison.email@example.com), Division of General Pediatrics, Department of Pediatrics, University of Minnesota, 717 Delaware Street SE, 3rd Floor, Room 370E, Minneapolis, MN 55414, USA; Nadia Maccabee-Ryaboy, University of Minnesota; Peter Scal, Andrew Wey, and Philippe Gaillard, University of Minnesota.